sleeping_kangaroo12 0.0.1 → 0.0.5
Sign up to get free protection for your applications and to get access to all the features.
- checksums.yaml +4 -4
- data/LICENSE.md +27 -0
- data/README.md +48 -53
- data/ext/Rakefile +12 -37
- data/ext/binding/sleeping_kangaroo12.c +1 -16
- data/ext/{xkcp → k12}/Makefile +0 -0
- data/ext/k12/Makefile.build +118 -0
- data/ext/k12/README.markdown +86 -0
- data/ext/k12/lib/ARMv8Asha3/KeccakP-1600-ARMv8Asha3.S +623 -0
- data/ext/k12/lib/ARMv8Asha3/KeccakP-1600-SnP.h +65 -0
- data/ext/k12/lib/ARMv8Asha3/KeccakP-1600-opt64.c +227 -0
- data/ext/{xkcp/lib/low/KeccakP-1600/compact → k12/lib/Inplace32BI}/KeccakP-1600-SnP.h +4 -9
- data/ext/{xkcp/lib/low/KeccakP-1600/plain-32bits-inplace → k12/lib/Inplace32BI}/KeccakP-1600-inplace32BI.c +65 -160
- data/ext/k12/lib/KangarooTwelve.c +332 -0
- data/ext/{xkcp/lib/high/KangarooTwelve → k12/lib}/KangarooTwelve.h +53 -16
- data/ext/{xkcp/lib/low/KeccakP-1600/AVX2 → k12/lib/Optimized64}/KeccakP-1600-AVX2.s +122 -558
- data/ext/k12/lib/Optimized64/KeccakP-1600-AVX512-plainC.c +241 -0
- data/ext/k12/lib/Optimized64/KeccakP-1600-AVX512.s +551 -0
- data/ext/k12/lib/Optimized64/KeccakP-1600-SnP.h +74 -0
- data/ext/{xkcp/lib/low/KeccakP-1600/common/KeccakP-1600-64.macros → k12/lib/Optimized64/KeccakP-1600-opt64.c} +447 -169
- data/ext/k12/lib/Optimized64/KeccakP-1600-runtimeDispatch.c +406 -0
- data/ext/k12/lib/Optimized64/KeccakP-1600-timesN-AVX2.c +419 -0
- data/ext/k12/lib/Optimized64/KeccakP-1600-timesN-AVX512.c +458 -0
- data/ext/k12/lib/Optimized64/KeccakP-1600-timesN-SSSE3.c +438 -0
- data/ext/{xkcp/lib/low/KeccakP-1600/plain-64bits → k12/lib/Plain64}/KeccakP-1600-SnP.h +14 -20
- data/ext/{xkcp/lib/low/KeccakP-1600/ref-64bits/KeccakP-1600-reference.h → k12/lib/Plain64/KeccakP-1600-plain64.c} +9 -8
- data/ext/{xkcp/lib/common → k12/lib}/align.h +3 -2
- data/ext/{xkcp/lib/common → k12/lib}/brg_endian.h +0 -0
- data/ext/{xkcp → k12}/support/Build/ExpandProducts.xsl +0 -0
- data/ext/{xkcp → k12}/support/Build/ToGlobalMakefile.xsl +0 -0
- data/ext/{xkcp → k12}/support/Build/ToOneTarget.xsl +0 -0
- data/ext/{xkcp → k12}/support/Build/ToTargetConfigFile.xsl +0 -0
- data/ext/{xkcp → k12}/support/Build/ToTargetMakefile.xsl +10 -16
- data/ext/{xkcp → k12}/support/Build/ToVCXProj.xsl +0 -0
- data/lib/sleeping_kangaroo12/binding.rb +2 -1
- data/lib/sleeping_kangaroo12/build/loader.rb +1 -0
- data/lib/sleeping_kangaroo12/build/platform.rb +1 -0
- data/lib/sleeping_kangaroo12/digest.rb +38 -4
- data/lib/sleeping_kangaroo12/version.rb +1 -1
- metadata +48 -288
- data/ext/config/xkcp.build +0 -17
- data/ext/xkcp/LICENSE +0 -1
- data/ext/xkcp/Makefile.build +0 -200
- data/ext/xkcp/README.markdown +0 -296
- data/ext/xkcp/lib/HighLevel.build +0 -143
- data/ext/xkcp/lib/LowLevel.build +0 -757
- data/ext/xkcp/lib/high/KangarooTwelve/KangarooTwelve.c +0 -301
- data/ext/xkcp/lib/high/Keccak/FIPS202/KeccakHash.c +0 -81
- data/ext/xkcp/lib/high/Keccak/FIPS202/KeccakHash.h +0 -125
- data/ext/xkcp/lib/high/Keccak/FIPS202/SimpleFIPS202.c +0 -48
- data/ext/xkcp/lib/high/Keccak/FIPS202/SimpleFIPS202.h +0 -79
- data/ext/xkcp/lib/high/Keccak/KeccakDuplex.c +0 -81
- data/ext/xkcp/lib/high/Keccak/KeccakDuplex.h +0 -73
- data/ext/xkcp/lib/high/Keccak/KeccakDuplex.inc +0 -195
- data/ext/xkcp/lib/high/Keccak/KeccakSponge.c +0 -111
- data/ext/xkcp/lib/high/Keccak/KeccakSponge.h +0 -76
- data/ext/xkcp/lib/high/Keccak/KeccakSponge.inc +0 -314
- data/ext/xkcp/lib/high/Keccak/PRG/KeccakPRG.c +0 -61
- data/ext/xkcp/lib/high/Keccak/PRG/KeccakPRG.h +0 -67
- data/ext/xkcp/lib/high/Keccak/PRG/KeccakPRG.inc +0 -128
- data/ext/xkcp/lib/high/Keccak/SP800-185/SP800-185.c +0 -93
- data/ext/xkcp/lib/high/Keccak/SP800-185/SP800-185.h +0 -599
- data/ext/xkcp/lib/high/Keccak/SP800-185/SP800-185.inc +0 -573
- data/ext/xkcp/lib/high/Ketje/Ketjev2.c +0 -87
- data/ext/xkcp/lib/high/Ketje/Ketjev2.h +0 -88
- data/ext/xkcp/lib/high/Ketje/Ketjev2.inc +0 -274
- data/ext/xkcp/lib/high/Keyak/Keyakv2.c +0 -132
- data/ext/xkcp/lib/high/Keyak/Keyakv2.h +0 -217
- data/ext/xkcp/lib/high/Keyak/Keyakv2.inc +0 -81
- data/ext/xkcp/lib/high/Keyak/Motorist.inc +0 -953
- data/ext/xkcp/lib/high/Kravatte/Kravatte.c +0 -533
- data/ext/xkcp/lib/high/Kravatte/Kravatte.h +0 -115
- data/ext/xkcp/lib/high/Kravatte/KravatteModes.c +0 -557
- data/ext/xkcp/lib/high/Kravatte/KravatteModes.h +0 -247
- data/ext/xkcp/lib/high/Xoodyak/Cyclist.h +0 -66
- data/ext/xkcp/lib/high/Xoodyak/Cyclist.inc +0 -336
- data/ext/xkcp/lib/high/Xoodyak/Xoodyak-parameters.h +0 -26
- data/ext/xkcp/lib/high/Xoodyak/Xoodyak.c +0 -55
- data/ext/xkcp/lib/high/Xoodyak/Xoodyak.h +0 -35
- data/ext/xkcp/lib/high/Xoofff/Xoofff.c +0 -634
- data/ext/xkcp/lib/high/Xoofff/Xoofff.h +0 -147
- data/ext/xkcp/lib/high/Xoofff/XoofffModes.c +0 -483
- data/ext/xkcp/lib/high/Xoofff/XoofffModes.h +0 -241
- data/ext/xkcp/lib/high/common/Phases.h +0 -25
- data/ext/xkcp/lib/low/KeccakP-1600/ARM/KeccakP-1600-SnP.h +0 -41
- data/ext/xkcp/lib/low/KeccakP-1600/ARM/KeccakP-1600-inplace-32bi-armv6m-le-armcc.s +0 -1666
- data/ext/xkcp/lib/low/KeccakP-1600/ARM/KeccakP-1600-inplace-32bi-armv6m-le-gcc.s +0 -1655
- data/ext/xkcp/lib/low/KeccakP-1600/ARM/KeccakP-1600-inplace-32bi-armv7a-le-armcc.s +0 -1268
- data/ext/xkcp/lib/low/KeccakP-1600/ARM/KeccakP-1600-inplace-32bi-armv7a-le-gcc.s +0 -1264
- data/ext/xkcp/lib/low/KeccakP-1600/ARM/KeccakP-1600-inplace-32bi-armv7m-le-armcc.s +0 -1178
- data/ext/xkcp/lib/low/KeccakP-1600/ARM/KeccakP-1600-inplace-32bi-armv7m-le-gcc.s +0 -1175
- data/ext/xkcp/lib/low/KeccakP-1600/ARM/KeccakP-1600-u1-32bi-armv6m-le-armcc.s +0 -1338
- data/ext/xkcp/lib/low/KeccakP-1600/ARM/KeccakP-1600-u1-32bi-armv6m-le-gcc.s +0 -1336
- data/ext/xkcp/lib/low/KeccakP-1600/ARM/KeccakP-1600-u2-32bi-armv6m-le-armcc.s +0 -1343
- data/ext/xkcp/lib/low/KeccakP-1600/ARM/KeccakP-1600-u2-32bi-armv6m-le-gcc.s +0 -1339
- data/ext/xkcp/lib/low/KeccakP-1600/ARMv7A-NEON/KeccakP-1600-SnP.h +0 -42
- data/ext/xkcp/lib/low/KeccakP-1600/ARMv7A-NEON/KeccakP-1600-armv7a-le-neon-armcc.s +0 -823
- data/ext/xkcp/lib/low/KeccakP-1600/ARMv7A-NEON/KeccakP-1600-armv7a-le-neon-gcc.s +0 -831
- data/ext/xkcp/lib/low/KeccakP-1600/ARMv8A/KeccakP-1600-SnP.h +0 -31
- data/ext/xkcp/lib/low/KeccakP-1600/ARMv8A/KeccakP-1600-armv8a-neon.s +0 -540
- data/ext/xkcp/lib/low/KeccakP-1600/AVR8/KeccakP-1600-SnP.h +0 -42
- data/ext/xkcp/lib/low/KeccakP-1600/AVR8/KeccakP-1600-avr8-compact.s +0 -733
- data/ext/xkcp/lib/low/KeccakP-1600/AVR8/KeccakP-1600-avr8-fast.s +0 -1121
- data/ext/xkcp/lib/low/KeccakP-1600/AVX2/KeccakP-1600-SnP.h +0 -52
- data/ext/xkcp/lib/low/KeccakP-1600/AVX512/C/KeccakP-1600-AVX512.c +0 -623
- data/ext/xkcp/lib/low/KeccakP-1600/AVX512/C/KeccakP-1600-SnP.h +0 -47
- data/ext/xkcp/lib/low/KeccakP-1600/AVX512/C/u12/KeccakP-1600-AVX512-config.h +0 -6
- data/ext/xkcp/lib/low/KeccakP-1600/AVX512/C/u6/KeccakP-1600-AVX512-config.h +0 -6
- data/ext/xkcp/lib/low/KeccakP-1600/AVX512/C/ua/KeccakP-1600-AVX512-config.h +0 -6
- data/ext/xkcp/lib/low/KeccakP-1600/AVX512/KeccakP-1600-AVX512.s +0 -1031
- data/ext/xkcp/lib/low/KeccakP-1600/AVX512/KeccakP-1600-SnP.h +0 -53
- data/ext/xkcp/lib/low/KeccakP-1600/XOP/KeccakP-1600-SnP.h +0 -44
- data/ext/xkcp/lib/low/KeccakP-1600/XOP/KeccakP-1600-XOP.c +0 -476
- data/ext/xkcp/lib/low/KeccakP-1600/XOP/u6/KeccakP-1600-XOP-config.h +0 -6
- data/ext/xkcp/lib/low/KeccakP-1600/XOP/ua/KeccakP-1600-XOP-config.h +0 -6
- data/ext/xkcp/lib/low/KeccakP-1600/common/KeccakP-1600-unrolling.macros +0 -305
- data/ext/xkcp/lib/low/KeccakP-1600/compact/KeccakP-1600-compact64.c +0 -420
- data/ext/xkcp/lib/low/KeccakP-1600/plain-32bits-inplace/KeccakP-1600-SnP.h +0 -43
- data/ext/xkcp/lib/low/KeccakP-1600/plain-64bits/KeccakP-1600-opt64.c +0 -565
- data/ext/xkcp/lib/low/KeccakP-1600/plain-64bits/lcu6/KeccakP-1600-opt64-config.h +0 -7
- data/ext/xkcp/lib/low/KeccakP-1600/plain-64bits/lcua/KeccakP-1600-opt64-config.h +0 -7
- data/ext/xkcp/lib/low/KeccakP-1600/plain-64bits/lcua-shld/KeccakP-1600-opt64-config.h +0 -8
- data/ext/xkcp/lib/low/KeccakP-1600/plain-64bits/u6/KeccakP-1600-opt64-config.h +0 -6
- data/ext/xkcp/lib/low/KeccakP-1600/plain-64bits/ua/KeccakP-1600-opt64-config.h +0 -6
- data/ext/xkcp/lib/low/KeccakP-1600/ref-32bits/KeccakP-1600-SnP.h +0 -44
- data/ext/xkcp/lib/low/KeccakP-1600/ref-32bits/KeccakP-1600-reference.h +0 -23
- data/ext/xkcp/lib/low/KeccakP-1600/ref-32bits/KeccakP-1600-reference32BI.c +0 -625
- data/ext/xkcp/lib/low/KeccakP-1600/ref-64bits/KeccakP-1600-SnP.h +0 -44
- data/ext/xkcp/lib/low/KeccakP-1600/ref-64bits/KeccakP-1600-reference.c +0 -440
- data/ext/xkcp/lib/low/KeccakP-1600/x86-64/KeccakP-1600-SnP.h +0 -42
- data/ext/xkcp/lib/low/KeccakP-1600/x86-64/KeccakP-1600-x86-64-gas.s +0 -1196
- data/ext/xkcp/lib/low/KeccakP-1600/x86-64/KeccakP-1600-x86-64-gas_Apple.s +0 -1124
- data/ext/xkcp/lib/low/KeccakP-1600/x86-64/KeccakP-1600-x86-64-shld-gas.s +0 -1196
- data/ext/xkcp/lib/low/KeccakP-1600-times2/ARMv7A-NEON/KeccakP-1600-inplace-pl2-armv7a-neon-le-armcc.s +0 -1392
- data/ext/xkcp/lib/low/KeccakP-1600-times2/ARMv7A-NEON/KeccakP-1600-inplace-pl2-armv7a-neon-le-gcc.s +0 -1394
- data/ext/xkcp/lib/low/KeccakP-1600-times2/ARMv7A-NEON/KeccakP-1600-times2-SnP.h +0 -42
- data/ext/xkcp/lib/low/KeccakP-1600-times2/AVX512/AVX512u12/SIMD512-2-config.h +0 -7
- data/ext/xkcp/lib/low/KeccakP-1600-times2/AVX512/AVX512u4/SIMD512-2-config.h +0 -7
- data/ext/xkcp/lib/low/KeccakP-1600-times2/AVX512/AVX512ufull/SIMD512-2-config.h +0 -7
- data/ext/xkcp/lib/low/KeccakP-1600-times2/AVX512/KeccakP-1600-times2-SIMD512.c +0 -850
- data/ext/xkcp/lib/low/KeccakP-1600-times2/AVX512/KeccakP-1600-times2-SnP.h +0 -51
- data/ext/xkcp/lib/low/KeccakP-1600-times2/SIMD128/KeccakP-1600-times2-SIMD128.c +0 -957
- data/ext/xkcp/lib/low/KeccakP-1600-times2/SIMD128/KeccakP-1600-times2-SnP.h +0 -49
- data/ext/xkcp/lib/low/KeccakP-1600-times2/SIMD128/SSSE3-u2/SIMD128-config.h +0 -8
- data/ext/xkcp/lib/low/KeccakP-1600-times2/SIMD128/SSSE3-ua/SIMD128-config.h +0 -8
- data/ext/xkcp/lib/low/KeccakP-1600-times2/SIMD128/XOP-u2/SIMD128-config.h +0 -9
- data/ext/xkcp/lib/low/KeccakP-1600-times2/SIMD128/XOP-ua/SIMD128-config.h +0 -9
- data/ext/xkcp/lib/low/KeccakP-1600-times2/fallback-on1/KeccakP-1600-times2-SnP.h +0 -45
- data/ext/xkcp/lib/low/KeccakP-1600-times2/fallback-on1/KeccakP-1600-times2-on1.c +0 -37
- data/ext/xkcp/lib/low/KeccakP-1600-times4/AVX2/KeccakP-1600-times4-SIMD256.c +0 -1321
- data/ext/xkcp/lib/low/KeccakP-1600-times4/AVX2/KeccakP-1600-times4-SnP.h +0 -55
- data/ext/xkcp/lib/low/KeccakP-1600-times4/AVX2/u12/SIMD256-config.h +0 -7
- data/ext/xkcp/lib/low/KeccakP-1600-times4/AVX2/u6/SIMD256-config.h +0 -7
- data/ext/xkcp/lib/low/KeccakP-1600-times4/AVX2/ua/SIMD256-config.h +0 -7
- data/ext/xkcp/lib/low/KeccakP-1600-times4/AVX512/AVX512u12/SIMD512-4-config.h +0 -7
- data/ext/xkcp/lib/low/KeccakP-1600-times4/AVX512/AVX512u4/SIMD512-4-config.h +0 -7
- data/ext/xkcp/lib/low/KeccakP-1600-times4/AVX512/AVX512ufull/SIMD512-4-config.h +0 -7
- data/ext/xkcp/lib/low/KeccakP-1600-times4/AVX512/KeccakP-1600-times4-SIMD512.c +0 -881
- data/ext/xkcp/lib/low/KeccakP-1600-times4/AVX512/KeccakP-1600-times4-SnP.h +0 -51
- data/ext/xkcp/lib/low/KeccakP-1600-times4/fallback-on1/KeccakP-1600-times4-SnP.h +0 -45
- data/ext/xkcp/lib/low/KeccakP-1600-times4/fallback-on1/KeccakP-1600-times4-on1.c +0 -37
- data/ext/xkcp/lib/low/KeccakP-1600-times4/fallback-on2/KeccakP-1600-times4-SnP.h +0 -45
- data/ext/xkcp/lib/low/KeccakP-1600-times4/fallback-on2/KeccakP-1600-times4-on2.c +0 -38
- data/ext/xkcp/lib/low/KeccakP-1600-times8/AVX512/KeccakP-1600-times8-SIMD512.c +0 -1615
- data/ext/xkcp/lib/low/KeccakP-1600-times8/AVX512/KeccakP-1600-times8-SnP.h +0 -57
- data/ext/xkcp/lib/low/KeccakP-1600-times8/AVX512/u12/SIMD512-config.h +0 -7
- data/ext/xkcp/lib/low/KeccakP-1600-times8/AVX512/u4/SIMD512-config.h +0 -7
- data/ext/xkcp/lib/low/KeccakP-1600-times8/AVX512/ua/SIMD512-config.h +0 -7
- data/ext/xkcp/lib/low/KeccakP-1600-times8/fallback-on1/KeccakP-1600-times8-SnP.h +0 -45
- data/ext/xkcp/lib/low/KeccakP-1600-times8/fallback-on1/KeccakP-1600-times8-on1.c +0 -37
- data/ext/xkcp/lib/low/KeccakP-1600-times8/fallback-on2/KeccakP-1600-times8-SnP.h +0 -45
- data/ext/xkcp/lib/low/KeccakP-1600-times8/fallback-on2/KeccakP-1600-times8-on2.c +0 -38
- data/ext/xkcp/lib/low/KeccakP-1600-times8/fallback-on4/KeccakP-1600-times8-SnP.h +0 -45
- data/ext/xkcp/lib/low/KeccakP-1600-times8/fallback-on4/KeccakP-1600-times8-on4.c +0 -38
- data/ext/xkcp/lib/low/KeccakP-200/ARM/KeccakP-200-SnP.h +0 -41
- data/ext/xkcp/lib/low/KeccakP-200/ARM/KeccakP-200-armv6m-le-armcc.s +0 -442
- data/ext/xkcp/lib/low/KeccakP-200/ARM/KeccakP-200-armv6m-le-gcc.s +0 -446
- data/ext/xkcp/lib/low/KeccakP-200/ARM/KeccakP-200-armv7m-le-armcc.s +0 -419
- data/ext/xkcp/lib/low/KeccakP-200/ARM/KeccakP-200-armv7m-le-gcc.s +0 -427
- data/ext/xkcp/lib/low/KeccakP-200/AVR8/KeccakP-200-SnP.h +0 -41
- data/ext/xkcp/lib/low/KeccakP-200/AVR8/KeccakP-200-avr8-fast.s +0 -647
- data/ext/xkcp/lib/low/KeccakP-200/compact/KeccakP-200-SnP.h +0 -39
- data/ext/xkcp/lib/low/KeccakP-200/compact/KeccakP-200-compact.c +0 -190
- data/ext/xkcp/lib/low/KeccakP-200/ref/KeccakP-200-SnP.h +0 -43
- data/ext/xkcp/lib/low/KeccakP-200/ref/KeccakP-200-reference.c +0 -412
- data/ext/xkcp/lib/low/KeccakP-200/ref/KeccakP-200-reference.h +0 -23
- data/ext/xkcp/lib/low/KeccakP-400/ARM/KeccakP-400-SnP.h +0 -41
- data/ext/xkcp/lib/low/KeccakP-400/ARM/KeccakP-400-armv6m-le-armcc.s +0 -454
- data/ext/xkcp/lib/low/KeccakP-400/ARM/KeccakP-400-armv6m-le-gcc.s +0 -458
- data/ext/xkcp/lib/low/KeccakP-400/ARM/KeccakP-400-armv7m-le-armcc.s +0 -455
- data/ext/xkcp/lib/low/KeccakP-400/ARM/KeccakP-400-armv7m-le-gcc.s +0 -458
- data/ext/xkcp/lib/low/KeccakP-400/AVR8/KeccakP-400-SnP.h +0 -41
- data/ext/xkcp/lib/low/KeccakP-400/AVR8/KeccakP-400-avr8-fast.s +0 -728
- data/ext/xkcp/lib/low/KeccakP-400/ref/KeccakP-400-SnP.h +0 -43
- data/ext/xkcp/lib/low/KeccakP-400/ref/KeccakP-400-reference.c +0 -414
- data/ext/xkcp/lib/low/KeccakP-400/ref/KeccakP-400-reference.h +0 -23
- data/ext/xkcp/lib/low/KeccakP-800/ARM/KeccakP-800-SnP.h +0 -42
- data/ext/xkcp/lib/low/KeccakP-800/ARM/KeccakP-800-u1-armv6m-le-armcc.s +0 -527
- data/ext/xkcp/lib/low/KeccakP-800/ARM/KeccakP-800-u1-armv6m-le-gcc.s +0 -533
- data/ext/xkcp/lib/low/KeccakP-800/ARM/KeccakP-800-u2-armv6m-le-armcc.s +0 -528
- data/ext/xkcp/lib/low/KeccakP-800/ARM/KeccakP-800-u2-armv6m-le-gcc.s +0 -534
- data/ext/xkcp/lib/low/KeccakP-800/ARM/KeccakP-800-u2-armv7a-le-armcc.s +0 -521
- data/ext/xkcp/lib/low/KeccakP-800/ARM/KeccakP-800-u2-armv7a-le-gcc.s +0 -527
- data/ext/xkcp/lib/low/KeccakP-800/ARM/KeccakP-800-u2-armv7m-le-armcc.s +0 -517
- data/ext/xkcp/lib/low/KeccakP-800/ARM/KeccakP-800-u2-armv7m-le-gcc.s +0 -523
- data/ext/xkcp/lib/low/KeccakP-800/ARM/KeccakP-800-uf-armv7m-le-armcc.s +0 -550
- data/ext/xkcp/lib/low/KeccakP-800/ARM/KeccakP-800-uf-armv7m-le-gcc.s +0 -556
- data/ext/xkcp/lib/low/KeccakP-800/ARMv8A/KeccakP-800-SnP.h +0 -32
- data/ext/xkcp/lib/low/KeccakP-800/ARMv8A/KeccakP-800-armv8a-neon.s +0 -432
- data/ext/xkcp/lib/low/KeccakP-800/AVR8/KeccakP-800-SnP.h +0 -42
- data/ext/xkcp/lib/low/KeccakP-800/AVR8/KeccakP-800-avr8-fast.s +0 -929
- data/ext/xkcp/lib/low/KeccakP-800/compact/KeccakP-800-SnP.h +0 -40
- data/ext/xkcp/lib/low/KeccakP-800/compact/KeccakP-800-compact.c +0 -244
- data/ext/xkcp/lib/low/KeccakP-800/plain/KeccakP-800-SnP.h +0 -46
- data/ext/xkcp/lib/low/KeccakP-800/plain/KeccakP-800-opt32-bis.macros +0 -184
- data/ext/xkcp/lib/low/KeccakP-800/plain/KeccakP-800-opt32.c +0 -454
- data/ext/xkcp/lib/low/KeccakP-800/plain/KeccakP-800-opt32.macros +0 -459
- data/ext/xkcp/lib/low/KeccakP-800/plain/KeccakP-800-unrolling-bis.macros +0 -83
- data/ext/xkcp/lib/low/KeccakP-800/plain/KeccakP-800-unrolling.macros +0 -88
- data/ext/xkcp/lib/low/KeccakP-800/plain/lcu2/KeccakP-800-opt32-config.h +0 -7
- data/ext/xkcp/lib/low/KeccakP-800/plain/lcua/KeccakP-800-opt32-config.h +0 -7
- data/ext/xkcp/lib/low/KeccakP-800/plain/u2/KeccakP-800-opt32-config.h +0 -7
- data/ext/xkcp/lib/low/KeccakP-800/plain/ua/KeccakP-800-opt32-config.h +0 -7
- data/ext/xkcp/lib/low/KeccakP-800/ref/KeccakP-800-SnP.h +0 -44
- data/ext/xkcp/lib/low/KeccakP-800/ref/KeccakP-800-reference.c +0 -437
- data/ext/xkcp/lib/low/KeccakP-800/ref/KeccakP-800-reference.h +0 -23
- data/ext/xkcp/lib/low/Ketje/OptimizedAsmARM/Ket.h +0 -57
- data/ext/xkcp/lib/low/Ketje/OptimizedAsmARM/KetjeJr-armv7m-le-armcc.s +0 -475
- data/ext/xkcp/lib/low/Ketje/OptimizedAsmARM/KetjeJr-armv7m-le-gcc.s +0 -480
- data/ext/xkcp/lib/low/Ketje/OptimizedAsmARM/KetjeSr-armv7m-le-armcc.s +0 -590
- data/ext/xkcp/lib/low/Ketje/OptimizedAsmARM/KetjeSr-armv7m-le-gcc.s +0 -590
- data/ext/xkcp/lib/low/Ketje/OptimizedLE/Ket.c +0 -126
- data/ext/xkcp/lib/low/Ketje/OptimizedLE/Ket.h +0 -68
- data/ext/xkcp/lib/low/Ketje/OptimizedLE/Ket.inc +0 -174
- data/ext/xkcp/lib/low/Ketje/SnP-compliant/Ket.c +0 -80
- data/ext/xkcp/lib/low/Ketje/SnP-compliant/Ket.h +0 -68
- data/ext/xkcp/lib/low/Ketje/SnP-compliant/Ket.inc +0 -142
- data/ext/xkcp/lib/low/Xoodoo/ARM/Xoodoo-SnP.h +0 -55
- data/ext/xkcp/lib/low/Xoodoo/ARM/Xoodoo-u1-armv6m-le-armcc.s +0 -1086
- data/ext/xkcp/lib/low/Xoodoo/ARM/Xoodoo-u1-armv6m-le-gcc.s +0 -1092
- data/ext/xkcp/lib/low/Xoodoo/ARM/Xoodoo-uf-armv6-le-armcc.s +0 -721
- data/ext/xkcp/lib/low/Xoodoo/ARM/Xoodoo-uf-armv6-le-gcc.s +0 -726
- data/ext/xkcp/lib/low/Xoodoo/ARM/Xoodoo-uf-armv7m-le-armcc.s +0 -723
- data/ext/xkcp/lib/low/Xoodoo/ARM/Xoodoo-uf-armv7m-le-gcc.s +0 -729
- data/ext/xkcp/lib/low/Xoodoo/ARM/Xoodyak-u1-armv6m-le-armcc.s +0 -1164
- data/ext/xkcp/lib/low/Xoodoo/ARM/Xoodyak-u1-armv6m-le-gcc.s +0 -1165
- data/ext/xkcp/lib/low/Xoodoo/ARM/Xoodyak-uf-armv6-le-armcc.s +0 -562
- data/ext/xkcp/lib/low/Xoodoo/ARM/Xoodyak-uf-armv6-le-gcc.s +0 -563
- data/ext/xkcp/lib/low/Xoodoo/ARM/Xoodyak-uf-armv7m-le-armcc.s +0 -563
- data/ext/xkcp/lib/low/Xoodoo/ARM/Xoodyak-uf-armv7m-le-gcc.s +0 -565
- data/ext/xkcp/lib/low/Xoodoo/ARMv7A-NEON/Xoodoo-SnP.h +0 -55
- data/ext/xkcp/lib/low/Xoodoo/ARMv7A-NEON/Xoodoo-uf-armv7a-neon-le-armcc.s +0 -476
- data/ext/xkcp/lib/low/Xoodoo/ARMv7A-NEON/Xoodoo-uf-armv7a-neon-le-gcc.s +0 -485
- data/ext/xkcp/lib/low/Xoodoo/ARMv7A-NEON/Xoodyak-uf-armv7a-neon-le-armcc.s +0 -362
- data/ext/xkcp/lib/low/Xoodoo/ARMv7A-NEON/Xoodyak-uf-armv7a-neon-le-gcc.s +0 -367
- data/ext/xkcp/lib/low/Xoodoo/AVR8/Xoodoo-SnP.h +0 -43
- data/ext/xkcp/lib/low/Xoodoo/AVR8/Xoodoo-avr8-u1.s +0 -1341
- data/ext/xkcp/lib/low/Xoodoo/AVX512/Xoodoo-SIMD512.c +0 -581
- data/ext/xkcp/lib/low/Xoodoo/AVX512/Xoodoo-SnP.h +0 -58
- data/ext/xkcp/lib/low/Xoodoo/AVX512/Xoodyak-full-block-SIMD512.c +0 -332
- data/ext/xkcp/lib/low/Xoodoo/SSE2/Xoodoo-SIMD128.c +0 -329
- data/ext/xkcp/lib/low/Xoodoo/SSE2/Xoodoo-SnP.h +0 -53
- data/ext/xkcp/lib/low/Xoodoo/SSE2/Xoodyak-full-block-SIMD128.c +0 -355
- data/ext/xkcp/lib/low/Xoodoo/Xoodoo.h +0 -79
- data/ext/xkcp/lib/low/Xoodoo/plain/Xoodoo-SnP.h +0 -56
- data/ext/xkcp/lib/low/Xoodoo/plain/Xoodoo-optimized.c +0 -399
- data/ext/xkcp/lib/low/Xoodoo/plain/Xoodyak-full-blocks.c +0 -127
- data/ext/xkcp/lib/low/Xoodoo/ref/Xoodoo-SnP.h +0 -43
- data/ext/xkcp/lib/low/Xoodoo/ref/Xoodoo-reference.c +0 -253
- data/ext/xkcp/lib/low/Xoodoo-times16/AVX512/Xoodoo-times16-SIMD512.c +0 -1044
- data/ext/xkcp/lib/low/Xoodoo-times16/AVX512/Xoodoo-times16-SnP.h +0 -49
- data/ext/xkcp/lib/low/Xoodoo-times16/fallback-on1/Xoodoo-times16-SnP.h +0 -45
- data/ext/xkcp/lib/low/Xoodoo-times16/fallback-on1/Xoodoo-times16-on1.c +0 -37
- data/ext/xkcp/lib/low/Xoodoo-times4/ARMv7A-NEON/Xoodoo-times4-ARMv7A.s +0 -1587
- data/ext/xkcp/lib/low/Xoodoo-times4/ARMv7A-NEON/Xoodoo-times4-SnP.h +0 -48
- data/ext/xkcp/lib/low/Xoodoo-times4/AVX512/Xoodoo-times4-SIMD512.c +0 -1202
- data/ext/xkcp/lib/low/Xoodoo-times4/AVX512/Xoodoo-times4-SnP.h +0 -48
- data/ext/xkcp/lib/low/Xoodoo-times4/SSSE3/Xoodoo-times4-SIMD128.c +0 -484
- data/ext/xkcp/lib/low/Xoodoo-times4/SSSE3/Xoodoo-times4-SnP.h +0 -44
- data/ext/xkcp/lib/low/Xoodoo-times4/fallback-on1/Xoodoo-times4-SnP.h +0 -45
- data/ext/xkcp/lib/low/Xoodoo-times4/fallback-on1/Xoodoo-times4-on1.c +0 -37
- data/ext/xkcp/lib/low/Xoodoo-times8/AVX2/Xoodoo-times8-SIMD256.c +0 -939
- data/ext/xkcp/lib/low/Xoodoo-times8/AVX2/Xoodoo-times8-SnP.h +0 -49
- data/ext/xkcp/lib/low/Xoodoo-times8/AVX512/Xoodoo-times8-SIMD512.c +0 -1216
- data/ext/xkcp/lib/low/Xoodoo-times8/AVX512/Xoodoo-times8-SnP.h +0 -48
- data/ext/xkcp/lib/low/Xoodoo-times8/fallback-on1/Xoodoo-times8-SnP.h +0 -45
- data/ext/xkcp/lib/low/Xoodoo-times8/fallback-on1/Xoodoo-times8-on1.c +0 -37
- data/ext/xkcp/lib/low/common/PlSnP-Fallback.inc +0 -290
- data/ext/xkcp/lib/low/common/SnP-Relaned.h +0 -141
- data/ext/xkcp/support/Kernel-PMU/Kernel-pmu.md +0 -133
- data/ext/xkcp/support/Kernel-PMU/Makefile +0 -8
- data/ext/xkcp/support/Kernel-PMU/enable_arm_pmu.c +0 -129
- data/ext/xkcp/support/Kernel-PMU/load-module +0 -1
- data/ext/xkcp/util/KeccakSum/KeccakSum.c +0 -394
- data/ext/xkcp/util/KeccakSum/base64.c +0 -86
- data/ext/xkcp/util/KeccakSum/base64.h +0 -12
@@ -0,0 +1,458 @@
|
|
1
|
+
/*
|
2
|
+
K12 based on the eXtended Keccak Code Package (XKCP)
|
3
|
+
https://github.com/XKCP/XKCP
|
4
|
+
|
5
|
+
The Keccak-p permutations, designed by Guido Bertoni, Joan Daemen, Michaël Peeters and Gilles Van Assche.
|
6
|
+
|
7
|
+
Implementation by Gilles Van Assche and Ronny Van Keer, hereby denoted as "the implementer".
|
8
|
+
|
9
|
+
For more information, feedback or questions, please refer to the Keccak Team website:
|
10
|
+
https://keccak.team/
|
11
|
+
|
12
|
+
To the extent possible under law, the implementer has waived all copyright
|
13
|
+
and related or neighboring rights to the source code in this file.
|
14
|
+
http://creativecommons.org/publicdomain/zero/1.0/
|
15
|
+
|
16
|
+
---
|
17
|
+
|
18
|
+
Please refer to the XKCP for more details.
|
19
|
+
*/
|
20
|
+
|
21
|
+
#include <stdint.h>
|
22
|
+
#include <emmintrin.h>
|
23
|
+
#include <immintrin.h>
|
24
|
+
#include "KeccakP-1600-SnP.h"
|
25
|
+
#include "align.h"
|
26
|
+
|
27
|
+
#define AVX512alignment 64
|
28
|
+
|
29
|
+
#define LOAD4_32(a,b,c,d) _mm_set_epi32((uint64_t)(a), (uint32_t)(b), (uint32_t)(c), (uint32_t)(d))
|
30
|
+
#define LOAD8_32(a,b,c,d,e,f,g,h) _mm256_set_epi32((uint64_t)(a), (uint32_t)(b), (uint32_t)(c), (uint32_t)(d), (uint32_t)(e), (uint32_t)(f), (uint32_t)(g), (uint32_t)(h))
|
31
|
+
#define LOAD_GATHER2_64(idx,p) _mm_i32gather_epi64( (const void*)(p), idx, 8)
|
32
|
+
#define LOAD_GATHER4_64(idx,p) _mm256_i32gather_epi64( (const void*)(p), idx, 8)
|
33
|
+
#define LOAD_GATHER8_64(idx,p) _mm512_i32gather_epi64( idx, (const void*)(p), 8)
|
34
|
+
#define STORE_SCATTER8_64(p,idx, v) _mm512_i32scatter_epi64( (void*)(p), idx, v, 8)
|
35
|
+
|
36
|
+
|
37
|
+
/* Keccak-p[1600]×2 */
|
38
|
+
|
39
|
+
#define XOR(a,b) _mm_xor_si128(a,b)
|
40
|
+
#define XOReq(a, b) a = _mm_xor_si128(a, b)
|
41
|
+
#define XOR3(a,b,c) _mm_ternarylogic_epi64(a,b,c,0x96)
|
42
|
+
#define XOR5(a,b,c,d,e) XOR3(XOR3(a,b,c),d,e)
|
43
|
+
#define ROL(a,offset) _mm_rol_epi64(a,offset)
|
44
|
+
#define Chi(a,b,c) _mm_ternarylogic_epi64(a,b,c,0xD2)
|
45
|
+
#define CONST_64(a) _mm_set1_epi64x(a)
|
46
|
+
#define LOAD6464(a, b) _mm_set_epi64x(a, b)
|
47
|
+
#define STORE128u(a, b) _mm_storeu_si128((__m128i *)&(a), b)
|
48
|
+
#define UNPACKL( a, b ) _mm_unpacklo_epi64((a), (b))
|
49
|
+
#define UNPACKH( a, b ) _mm_unpackhi_epi64((a), (b))
|
50
|
+
#define ZERO() _mm_setzero_si128()
|
51
|
+
|
52
|
+
static ALIGN(AVX512alignment) const uint64_t KeccakP1600RoundConstants[24] = {
|
53
|
+
0x0000000000000001ULL,
|
54
|
+
0x0000000000008082ULL,
|
55
|
+
0x800000000000808aULL,
|
56
|
+
0x8000000080008000ULL,
|
57
|
+
0x000000000000808bULL,
|
58
|
+
0x0000000080000001ULL,
|
59
|
+
0x8000000080008081ULL,
|
60
|
+
0x8000000000008009ULL,
|
61
|
+
0x000000000000008aULL,
|
62
|
+
0x0000000000000088ULL,
|
63
|
+
0x0000000080008009ULL,
|
64
|
+
0x000000008000000aULL,
|
65
|
+
0x000000008000808bULL,
|
66
|
+
0x800000000000008bULL,
|
67
|
+
0x8000000000008089ULL,
|
68
|
+
0x8000000000008003ULL,
|
69
|
+
0x8000000000008002ULL,
|
70
|
+
0x8000000000000080ULL,
|
71
|
+
0x000000000000800aULL,
|
72
|
+
0x800000008000000aULL,
|
73
|
+
0x8000000080008081ULL,
|
74
|
+
0x8000000000008080ULL,
|
75
|
+
0x0000000080000001ULL,
|
76
|
+
0x8000000080008008ULL};
|
77
|
+
|
78
|
+
#define KeccakP_DeclareVars(type) \
|
79
|
+
type _Ba, _Be, _Bi, _Bo, _Bu; \
|
80
|
+
type _Da, _De, _Di, _Do, _Du; \
|
81
|
+
type _ba, _be, _bi, _bo, _bu; \
|
82
|
+
type _ga, _ge, _gi, _go, _gu; \
|
83
|
+
type _ka, _ke, _ki, _ko, _ku; \
|
84
|
+
type _ma, _me, _mi, _mo, _mu; \
|
85
|
+
type _sa, _se, _si, _so, _su
|
86
|
+
|
87
|
+
#define KeccakP_ThetaRhoPiChi( _L1, _L2, _L3, _L4, _L5, _Bb1, _Bb2, _Bb3, _Bb4, _Bb5, _Rr1, _Rr2, _Rr3, _Rr4, _Rr5 ) \
|
88
|
+
_Bb1 = XOR(_L1, _Da); \
|
89
|
+
_Bb2 = XOR(_L2, _De); \
|
90
|
+
_Bb3 = XOR(_L3, _Di); \
|
91
|
+
_Bb4 = XOR(_L4, _Do); \
|
92
|
+
_Bb5 = XOR(_L5, _Du); \
|
93
|
+
if (_Rr1 != 0) _Bb1 = ROL(_Bb1, _Rr1); \
|
94
|
+
_Bb2 = ROL(_Bb2, _Rr2); \
|
95
|
+
_Bb3 = ROL(_Bb3, _Rr3); \
|
96
|
+
_Bb4 = ROL(_Bb4, _Rr4); \
|
97
|
+
_Bb5 = ROL(_Bb5, _Rr5); \
|
98
|
+
_L1 = Chi( _Ba, _Be, _Bi); \
|
99
|
+
_L2 = Chi( _Be, _Bi, _Bo); \
|
100
|
+
_L3 = Chi( _Bi, _Bo, _Bu); \
|
101
|
+
_L4 = Chi( _Bo, _Bu, _Ba); \
|
102
|
+
_L5 = Chi( _Bu, _Ba, _Be);
|
103
|
+
|
104
|
+
#define KeccakP_ThetaRhoPiChiIota0( _L1, _L2, _L3, _L4, _L5, _rc ) \
|
105
|
+
_Ba = XOR5( _ba, _ga, _ka, _ma, _sa ); /* Theta effect */ \
|
106
|
+
_Be = XOR5( _be, _ge, _ke, _me, _se ); \
|
107
|
+
_Bi = XOR5( _bi, _gi, _ki, _mi, _si ); \
|
108
|
+
_Bo = XOR5( _bo, _go, _ko, _mo, _so ); \
|
109
|
+
_Bu = XOR5( _bu, _gu, _ku, _mu, _su ); \
|
110
|
+
_Da = ROL( _Be, 1 ); \
|
111
|
+
_De = ROL( _Bi, 1 ); \
|
112
|
+
_Di = ROL( _Bo, 1 ); \
|
113
|
+
_Do = ROL( _Bu, 1 ); \
|
114
|
+
_Du = ROL( _Ba, 1 ); \
|
115
|
+
_Da = XOR( _Da, _Bu ); \
|
116
|
+
_De = XOR( _De, _Ba ); \
|
117
|
+
_Di = XOR( _Di, _Be ); \
|
118
|
+
_Do = XOR( _Do, _Bi ); \
|
119
|
+
_Du = XOR( _Du, _Bo ); \
|
120
|
+
KeccakP_ThetaRhoPiChi( _L1, _L2, _L3, _L4, _L5, _Ba, _Be, _Bi, _Bo, _Bu, 0, 44, 43, 21, 14 ); \
|
121
|
+
_L1 = XOR(_L1, _rc) /* Iota */
|
122
|
+
|
123
|
+
#define KeccakP_ThetaRhoPiChi1( _L1, _L2, _L3, _L4, _L5 ) \
|
124
|
+
KeccakP_ThetaRhoPiChi( _L1, _L2, _L3, _L4, _L5, _Bi, _Bo, _Bu, _Ba, _Be, 3, 45, 61, 28, 20 )
|
125
|
+
|
126
|
+
#define KeccakP_ThetaRhoPiChi2( _L1, _L2, _L3, _L4, _L5 ) \
|
127
|
+
KeccakP_ThetaRhoPiChi( _L1, _L2, _L3, _L4, _L5, _Bu, _Ba, _Be, _Bi, _Bo, 18, 1, 6, 25, 8 )
|
128
|
+
|
129
|
+
#define KeccakP_ThetaRhoPiChi3( _L1, _L2, _L3, _L4, _L5 ) \
|
130
|
+
KeccakP_ThetaRhoPiChi( _L1, _L2, _L3, _L4, _L5, _Be, _Bi, _Bo, _Bu, _Ba, 36, 10, 15, 56, 27 )
|
131
|
+
|
132
|
+
#define KeccakP_ThetaRhoPiChi4( _L1, _L2, _L3, _L4, _L5 ) \
|
133
|
+
KeccakP_ThetaRhoPiChi( _L1, _L2, _L3, _L4, _L5, _Bo, _Bu, _Ba, _Be, _Bi, 41, 2, 62, 55, 39 )
|
134
|
+
|
135
|
+
#define KeccakP_4rounds( i ) \
|
136
|
+
KeccakP_ThetaRhoPiChiIota0(_ba, _ge, _ki, _mo, _su, CONST_64(KeccakP1600RoundConstants[i]) ); \
|
137
|
+
KeccakP_ThetaRhoPiChi1( _ka, _me, _si, _bo, _gu ); \
|
138
|
+
KeccakP_ThetaRhoPiChi2( _sa, _be, _gi, _ko, _mu ); \
|
139
|
+
KeccakP_ThetaRhoPiChi3( _ga, _ke, _mi, _so, _bu ); \
|
140
|
+
KeccakP_ThetaRhoPiChi4( _ma, _se, _bi, _go, _ku ); \
|
141
|
+
\
|
142
|
+
KeccakP_ThetaRhoPiChiIota0(_ba, _me, _gi, _so, _ku, CONST_64(KeccakP1600RoundConstants[i+1]) ); \
|
143
|
+
KeccakP_ThetaRhoPiChi1( _sa, _ke, _bi, _mo, _gu ); \
|
144
|
+
KeccakP_ThetaRhoPiChi2( _ma, _ge, _si, _ko, _bu ); \
|
145
|
+
KeccakP_ThetaRhoPiChi3( _ka, _be, _mi, _go, _su ); \
|
146
|
+
KeccakP_ThetaRhoPiChi4( _ga, _se, _ki, _bo, _mu ); \
|
147
|
+
\
|
148
|
+
KeccakP_ThetaRhoPiChiIota0(_ba, _ke, _si, _go, _mu, CONST_64(KeccakP1600RoundConstants[i+2]) ); \
|
149
|
+
KeccakP_ThetaRhoPiChi1( _ma, _be, _ki, _so, _gu ); \
|
150
|
+
KeccakP_ThetaRhoPiChi2( _ga, _me, _bi, _ko, _su ); \
|
151
|
+
KeccakP_ThetaRhoPiChi3( _sa, _ge, _mi, _bo, _ku ); \
|
152
|
+
KeccakP_ThetaRhoPiChi4( _ka, _se, _gi, _mo, _bu ); \
|
153
|
+
\
|
154
|
+
KeccakP_ThetaRhoPiChiIota0(_ba, _be, _bi, _bo, _bu, CONST_64(KeccakP1600RoundConstants[i+3]) ); \
|
155
|
+
KeccakP_ThetaRhoPiChi1( _ga, _ge, _gi, _go, _gu ); \
|
156
|
+
KeccakP_ThetaRhoPiChi2( _ka, _ke, _ki, _ko, _ku ); \
|
157
|
+
KeccakP_ThetaRhoPiChi3( _ma, _me, _mi, _mo, _mu ); \
|
158
|
+
KeccakP_ThetaRhoPiChi4( _sa, _se, _si, _so, _su )
|
159
|
+
|
160
|
+
#define rounds12 \
|
161
|
+
KeccakP_4rounds( 12 ); \
|
162
|
+
KeccakP_4rounds( 16 ); \
|
163
|
+
KeccakP_4rounds( 20 )
|
164
|
+
|
165
|
+
#define initializeState(X) \
|
166
|
+
X##ba = ZERO(); \
|
167
|
+
X##be = ZERO(); \
|
168
|
+
X##bi = ZERO(); \
|
169
|
+
X##bo = ZERO(); \
|
170
|
+
X##bu = ZERO(); \
|
171
|
+
X##ga = ZERO(); \
|
172
|
+
X##ge = ZERO(); \
|
173
|
+
X##gi = ZERO(); \
|
174
|
+
X##go = ZERO(); \
|
175
|
+
X##gu = ZERO(); \
|
176
|
+
X##ka = ZERO(); \
|
177
|
+
X##ke = ZERO(); \
|
178
|
+
X##ki = ZERO(); \
|
179
|
+
X##ko = ZERO(); \
|
180
|
+
X##ku = ZERO(); \
|
181
|
+
X##ma = ZERO(); \
|
182
|
+
X##me = ZERO(); \
|
183
|
+
X##mi = ZERO(); \
|
184
|
+
X##mo = ZERO(); \
|
185
|
+
X##mu = ZERO(); \
|
186
|
+
X##sa = ZERO(); \
|
187
|
+
X##se = ZERO(); \
|
188
|
+
X##si = ZERO(); \
|
189
|
+
X##so = ZERO(); \
|
190
|
+
X##su = ZERO(); \
|
191
|
+
|
192
|
+
#define XORdata16(X, data0, data1) \
|
193
|
+
XOReq(X##ba, LOAD6464((data1)[ 0], (data0)[ 0])); \
|
194
|
+
XOReq(X##be, LOAD6464((data1)[ 1], (data0)[ 1])); \
|
195
|
+
XOReq(X##bi, LOAD6464((data1)[ 2], (data0)[ 2])); \
|
196
|
+
XOReq(X##bo, LOAD6464((data1)[ 3], (data0)[ 3])); \
|
197
|
+
XOReq(X##bu, LOAD6464((data1)[ 4], (data0)[ 4])); \
|
198
|
+
XOReq(X##ga, LOAD6464((data1)[ 5], (data0)[ 5])); \
|
199
|
+
XOReq(X##ge, LOAD6464((data1)[ 6], (data0)[ 6])); \
|
200
|
+
XOReq(X##gi, LOAD6464((data1)[ 7], (data0)[ 7])); \
|
201
|
+
XOReq(X##go, LOAD6464((data1)[ 8], (data0)[ 8])); \
|
202
|
+
XOReq(X##gu, LOAD6464((data1)[ 9], (data0)[ 9])); \
|
203
|
+
XOReq(X##ka, LOAD6464((data1)[10], (data0)[10])); \
|
204
|
+
XOReq(X##ke, LOAD6464((data1)[11], (data0)[11])); \
|
205
|
+
XOReq(X##ki, LOAD6464((data1)[12], (data0)[12])); \
|
206
|
+
XOReq(X##ko, LOAD6464((data1)[13], (data0)[13])); \
|
207
|
+
XOReq(X##ku, LOAD6464((data1)[14], (data0)[14])); \
|
208
|
+
XOReq(X##ma, LOAD6464((data1)[15], (data0)[15])); \
|
209
|
+
|
210
|
+
#define XORdata21(X, data0, data1) \
|
211
|
+
XORdata16(X, data0, data1) \
|
212
|
+
XOReq(X##me, LOAD6464((data1)[16], (data0)[16])); \
|
213
|
+
XOReq(X##mi, LOAD6464((data1)[17], (data0)[17])); \
|
214
|
+
XOReq(X##mo, LOAD6464((data1)[18], (data0)[18])); \
|
215
|
+
XOReq(X##mu, LOAD6464((data1)[19], (data0)[19])); \
|
216
|
+
XOReq(X##sa, LOAD6464((data1)[20], (data0)[20])); \
|
217
|
+
|
218
|
+
#define chunkSize 8192
|
219
|
+
#define rateInBytes (21*8)
|
220
|
+
|
221
|
+
void KangarooTwelve_AVX512_Process2Leaves(const unsigned char *input, unsigned char *output)
|
222
|
+
{
|
223
|
+
KeccakP_DeclareVars(__m128i);
|
224
|
+
unsigned int j;
|
225
|
+
|
226
|
+
initializeState(_);
|
227
|
+
|
228
|
+
for(j = 0; j < (chunkSize - rateInBytes); j += rateInBytes) {
|
229
|
+
XORdata21(_, (const uint64_t *)input, (const uint64_t *)(input+chunkSize));
|
230
|
+
rounds12
|
231
|
+
input += rateInBytes;
|
232
|
+
}
|
233
|
+
|
234
|
+
XORdata16(_, (const uint64_t *)input, (const uint64_t *)(input+chunkSize));
|
235
|
+
XOReq(_me, CONST_64(0x0BULL));
|
236
|
+
XOReq(_sa, CONST_64(0x8000000000000000ULL));
|
237
|
+
rounds12
|
238
|
+
|
239
|
+
STORE128u( *(__m128i*)&(output[ 0]), UNPACKL( _ba, _be ) );
|
240
|
+
STORE128u( *(__m128i*)&(output[16]), UNPACKL( _bi, _bo ) );
|
241
|
+
STORE128u( *(__m128i*)&(output[32]), UNPACKH( _ba, _be ) );
|
242
|
+
STORE128u( *(__m128i*)&(output[48]), UNPACKH( _bi, _bo ) );
|
243
|
+
}
|
244
|
+
|
245
|
+
#undef XOR
|
246
|
+
#undef XOReq
|
247
|
+
#undef XOR3
|
248
|
+
#undef XOR5
|
249
|
+
#undef ROL
|
250
|
+
#undef Chi
|
251
|
+
#undef CONST_64
|
252
|
+
#undef LOAD6464
|
253
|
+
#undef STORE128u
|
254
|
+
#undef UNPACKL
|
255
|
+
#undef UNPACKH
|
256
|
+
#undef ZERO
|
257
|
+
#undef XORdata16
|
258
|
+
#undef XORdata21
|
259
|
+
|
260
|
+
|
261
|
+
/* Keccak-p[1600]×4 */
|
262
|
+
|
263
|
+
#define XOR(a,b) _mm256_xor_si256(a,b)
|
264
|
+
#define XOReq(a,b) a = _mm256_xor_si256(a,b)
|
265
|
+
#define XOR3(a,b,c) _mm256_ternarylogic_epi64(a,b,c,0x96)
|
266
|
+
#define XOR5(a,b,c,d,e) XOR3(XOR3(a,b,c),d,e)
|
267
|
+
#define XOR512(a,b) _mm512_xor_si512(a,b)
|
268
|
+
#define ROL(a,offset) _mm256_rol_epi64(a,offset)
|
269
|
+
#define Chi(a,b,c) _mm256_ternarylogic_epi64(a,b,c,0xD2)
|
270
|
+
#define CONST_64(a) _mm256_set1_epi64x(a)
|
271
|
+
#define ZERO() _mm256_setzero_si256()
|
272
|
+
#define LOAD4_64(a, b, c, d) _mm256_set_epi64x((uint64_t)(a), (uint64_t)(b), (uint64_t)(c), (uint64_t)(d))
|
273
|
+
|
274
|
+
#define XORdata16(X, data0, data1, data2, data3) \
|
275
|
+
XOReq(X##ba, LOAD4_64((data3)[ 0], (data2)[ 0], (data1)[ 0], (data0)[ 0])); \
|
276
|
+
XOReq(X##be, LOAD4_64((data3)[ 1], (data2)[ 1], (data1)[ 1], (data0)[ 1])); \
|
277
|
+
XOReq(X##bi, LOAD4_64((data3)[ 2], (data2)[ 2], (data1)[ 2], (data0)[ 2])); \
|
278
|
+
XOReq(X##bo, LOAD4_64((data3)[ 3], (data2)[ 3], (data1)[ 3], (data0)[ 3])); \
|
279
|
+
XOReq(X##bu, LOAD4_64((data3)[ 4], (data2)[ 4], (data1)[ 4], (data0)[ 4])); \
|
280
|
+
XOReq(X##ga, LOAD4_64((data3)[ 5], (data2)[ 5], (data1)[ 5], (data0)[ 5])); \
|
281
|
+
XOReq(X##ge, LOAD4_64((data3)[ 6], (data2)[ 6], (data1)[ 6], (data0)[ 6])); \
|
282
|
+
XOReq(X##gi, LOAD4_64((data3)[ 7], (data2)[ 7], (data1)[ 7], (data0)[ 7])); \
|
283
|
+
XOReq(X##go, LOAD4_64((data3)[ 8], (data2)[ 8], (data1)[ 8], (data0)[ 8])); \
|
284
|
+
XOReq(X##gu, LOAD4_64((data3)[ 9], (data2)[ 9], (data1)[ 9], (data0)[ 9])); \
|
285
|
+
XOReq(X##ka, LOAD4_64((data3)[10], (data2)[10], (data1)[10], (data0)[10])); \
|
286
|
+
XOReq(X##ke, LOAD4_64((data3)[11], (data2)[11], (data1)[11], (data0)[11])); \
|
287
|
+
XOReq(X##ki, LOAD4_64((data3)[12], (data2)[12], (data1)[12], (data0)[12])); \
|
288
|
+
XOReq(X##ko, LOAD4_64((data3)[13], (data2)[13], (data1)[13], (data0)[13])); \
|
289
|
+
XOReq(X##ku, LOAD4_64((data3)[14], (data2)[14], (data1)[14], (data0)[14])); \
|
290
|
+
XOReq(X##ma, LOAD4_64((data3)[15], (data2)[15], (data1)[15], (data0)[15])); \
|
291
|
+
|
292
|
+
#define XORdata21(X, data0, data1, data2, data3) \
|
293
|
+
XORdata16(X, data0, data1, data2, data3) \
|
294
|
+
XOReq(X##me, LOAD4_64((data3)[16], (data2)[16], (data1)[16], (data0)[16])); \
|
295
|
+
XOReq(X##mi, LOAD4_64((data3)[17], (data2)[17], (data1)[17], (data0)[17])); \
|
296
|
+
XOReq(X##mo, LOAD4_64((data3)[18], (data2)[18], (data1)[18], (data0)[18])); \
|
297
|
+
XOReq(X##mu, LOAD4_64((data3)[19], (data2)[19], (data1)[19], (data0)[19])); \
|
298
|
+
XOReq(X##sa, LOAD4_64((data3)[20], (data2)[20], (data1)[20], (data0)[20])); \
|
299
|
+
|
300
|
+
void KangarooTwelve_AVX512_Process4Leaves(const unsigned char *input, unsigned char *output)
|
301
|
+
{
|
302
|
+
KeccakP_DeclareVars(__m256i);
|
303
|
+
unsigned int j;
|
304
|
+
|
305
|
+
initializeState(_);
|
306
|
+
|
307
|
+
for(j = 0; j < (chunkSize - rateInBytes); j += rateInBytes) {
|
308
|
+
XORdata21(_, (const uint64_t *)input, (const uint64_t *)(input+chunkSize), (const uint64_t *)(input+2*chunkSize), (const uint64_t *)(input+3*chunkSize));
|
309
|
+
rounds12
|
310
|
+
input += rateInBytes;
|
311
|
+
}
|
312
|
+
|
313
|
+
XORdata16(_, (const uint64_t *)input, (const uint64_t *)(input+chunkSize), (const uint64_t *)(input+2*chunkSize), (const uint64_t *)(input+3*chunkSize));
|
314
|
+
XOReq(_me, CONST_64(0x0BULL));
|
315
|
+
XOReq(_sa, CONST_64(0x8000000000000000ULL));
|
316
|
+
rounds12
|
317
|
+
|
318
|
+
#define STORE256u(a, b) _mm256_storeu_si256((__m256i *)&(a), b)
|
319
|
+
#define UNPACKL( a, b ) _mm256_unpacklo_epi64((a), (b))
|
320
|
+
#define UNPACKH( a, b ) _mm256_unpackhi_epi64((a), (b))
|
321
|
+
#define PERM128( a, b, c ) _mm256_permute2f128_si256(a, b, c)
|
322
|
+
{
|
323
|
+
__m256i lanesL01, lanesL23, lanesH01, lanesH23;
|
324
|
+
|
325
|
+
lanesL01 = UNPACKL( _ba, _be );
|
326
|
+
lanesH01 = UNPACKH( _ba, _be );
|
327
|
+
lanesL23 = UNPACKL( _bi, _bo );
|
328
|
+
lanesH23 = UNPACKH( _bi, _bo );
|
329
|
+
STORE256u( output[ 0], PERM128( lanesL01, lanesL23, 0x20 ) );
|
330
|
+
STORE256u( output[32], PERM128( lanesH01, lanesH23, 0x20 ) );
|
331
|
+
STORE256u( output[64], PERM128( lanesL01, lanesL23, 0x31 ) );
|
332
|
+
STORE256u( output[96], PERM128( lanesH01, lanesH23, 0x31 ) );
|
333
|
+
}
|
334
|
+
/* TODO: check if something like this would be better:
|
335
|
+
index512 = LOAD8_32(3*laneOffset+1, 2*laneOffset+1, 1*laneOffset+1, 0*laneOffset+1, 3*laneOffset, 2*laneOffset, 1*laneOffset, 0*laneOffset);
|
336
|
+
STORE_SCATTER8_64(dataAsLanes+0, index512, stateAsLanes512[0/2]);
|
337
|
+
STORE_SCATTER8_64(dataAsLanes+2, index512, stateAsLanes512[2/2]);
|
338
|
+
*/
|
339
|
+
}
|
340
|
+
|
341
|
+
#undef XOR
|
342
|
+
#undef XOReq
|
343
|
+
#undef XOR3
|
344
|
+
#undef XOR5
|
345
|
+
#undef XOR512
|
346
|
+
#undef ROL
|
347
|
+
#undef Chi
|
348
|
+
#undef CONST_64
|
349
|
+
#undef ZERO
|
350
|
+
#undef LOAD4_64
|
351
|
+
#undef XORdata16
|
352
|
+
#undef XORdata21
|
353
|
+
|
354
|
+
|
355
|
+
/* Keccak-p[1600]×8 */
|
356
|
+
|
357
|
+
#define XOR(a,b) _mm512_xor_si512(a,b)
|
358
|
+
#define XOReq(a,b) a = _mm512_xor_si512(a,b)
|
359
|
+
#define XOR3(a,b,c) _mm512_ternarylogic_epi64(a,b,c,0x96)
|
360
|
+
#define XOR5(a,b,c,d,e) XOR3(XOR3(a,b,c),d,e)
|
361
|
+
#define XOReq512(a, b) a = XOR(a,b)
|
362
|
+
#define ROL(a,offset) _mm512_rol_epi64(a,offset)
|
363
|
+
#define Chi(a,b,c) _mm512_ternarylogic_epi64(a,b,c,0xD2)
|
364
|
+
#define CONST_64(a) _mm512_set1_epi64(a)
|
365
|
+
#define ZERO() _mm512_setzero_si512()
|
366
|
+
#define LOAD(p) _mm512_loadu_si512(p)
|
367
|
+
|
368
|
+
#define LoadAndTranspose8(dataAsLanes, offset) \
|
369
|
+
t0 = LOAD((dataAsLanes) + (offset) + 0*chunkSize/8); \
|
370
|
+
t1 = LOAD((dataAsLanes) + (offset) + 1*chunkSize/8); \
|
371
|
+
t2 = LOAD((dataAsLanes) + (offset) + 2*chunkSize/8); \
|
372
|
+
t3 = LOAD((dataAsLanes) + (offset) + 3*chunkSize/8); \
|
373
|
+
t4 = LOAD((dataAsLanes) + (offset) + 4*chunkSize/8); \
|
374
|
+
t5 = LOAD((dataAsLanes) + (offset) + 5*chunkSize/8); \
|
375
|
+
t6 = LOAD((dataAsLanes) + (offset) + 6*chunkSize/8); \
|
376
|
+
t7 = LOAD((dataAsLanes) + (offset) + 7*chunkSize/8); \
|
377
|
+
r0 = _mm512_unpacklo_epi64(t0, t1); \
|
378
|
+
r1 = _mm512_unpackhi_epi64(t0, t1); \
|
379
|
+
r2 = _mm512_unpacklo_epi64(t2, t3); \
|
380
|
+
r3 = _mm512_unpackhi_epi64(t2, t3); \
|
381
|
+
r4 = _mm512_unpacklo_epi64(t4, t5); \
|
382
|
+
r5 = _mm512_unpackhi_epi64(t4, t5); \
|
383
|
+
r6 = _mm512_unpacklo_epi64(t6, t7); \
|
384
|
+
r7 = _mm512_unpackhi_epi64(t6, t7); \
|
385
|
+
t0 = _mm512_shuffle_i32x4(r0, r2, 0x88); \
|
386
|
+
t1 = _mm512_shuffle_i32x4(r1, r3, 0x88); \
|
387
|
+
t2 = _mm512_shuffle_i32x4(r0, r2, 0xdd); \
|
388
|
+
t3 = _mm512_shuffle_i32x4(r1, r3, 0xdd); \
|
389
|
+
t4 = _mm512_shuffle_i32x4(r4, r6, 0x88); \
|
390
|
+
t5 = _mm512_shuffle_i32x4(r5, r7, 0x88); \
|
391
|
+
t6 = _mm512_shuffle_i32x4(r4, r6, 0xdd); \
|
392
|
+
t7 = _mm512_shuffle_i32x4(r5, r7, 0xdd); \
|
393
|
+
r0 = _mm512_shuffle_i32x4(t0, t4, 0x88); \
|
394
|
+
r1 = _mm512_shuffle_i32x4(t1, t5, 0x88); \
|
395
|
+
r2 = _mm512_shuffle_i32x4(t2, t6, 0x88); \
|
396
|
+
r3 = _mm512_shuffle_i32x4(t3, t7, 0x88); \
|
397
|
+
r4 = _mm512_shuffle_i32x4(t0, t4, 0xdd); \
|
398
|
+
r5 = _mm512_shuffle_i32x4(t1, t5, 0xdd); \
|
399
|
+
r6 = _mm512_shuffle_i32x4(t2, t6, 0xdd); \
|
400
|
+
r7 = _mm512_shuffle_i32x4(t3, t7, 0xdd); \
|
401
|
+
|
402
|
+
#define XORdata16(X, index, dataAsLanes) \
|
403
|
+
LoadAndTranspose8(dataAsLanes, 0) \
|
404
|
+
XOReq(X##ba, r0); \
|
405
|
+
XOReq(X##be, r1); \
|
406
|
+
XOReq(X##bi, r2); \
|
407
|
+
XOReq(X##bo, r3); \
|
408
|
+
XOReq(X##bu, r4); \
|
409
|
+
XOReq(X##ga, r5); \
|
410
|
+
XOReq(X##ge, r6); \
|
411
|
+
XOReq(X##gi, r7); \
|
412
|
+
LoadAndTranspose8(dataAsLanes, 8) \
|
413
|
+
XOReq(X##go, r0); \
|
414
|
+
XOReq(X##gu, r1); \
|
415
|
+
XOReq(X##ka, r2); \
|
416
|
+
XOReq(X##ke, r3); \
|
417
|
+
XOReq(X##ki, r4); \
|
418
|
+
XOReq(X##ko, r5); \
|
419
|
+
XOReq(X##ku, r6); \
|
420
|
+
XOReq(X##ma, r7); \
|
421
|
+
|
422
|
+
#define XORdata21(X, index, dataAsLanes) \
|
423
|
+
XORdata16(X, index, dataAsLanes) \
|
424
|
+
XOReq(X##me, LOAD_GATHER8_64(index, (dataAsLanes) + 16)); \
|
425
|
+
XOReq(X##mi, LOAD_GATHER8_64(index, (dataAsLanes) + 17)); \
|
426
|
+
XOReq(X##mo, LOAD_GATHER8_64(index, (dataAsLanes) + 18)); \
|
427
|
+
XOReq(X##mu, LOAD_GATHER8_64(index, (dataAsLanes) + 19)); \
|
428
|
+
XOReq(X##sa, LOAD_GATHER8_64(index, (dataAsLanes) + 20)); \
|
429
|
+
|
430
|
+
void KangarooTwelve_AVX512_Process8Leaves(const unsigned char *input, unsigned char *output)
|
431
|
+
{
|
432
|
+
KeccakP_DeclareVars(__m512i);
|
433
|
+
unsigned int j;
|
434
|
+
const uint64_t *outputAsLanes = (const uint64_t *)output;
|
435
|
+
__m256i index;
|
436
|
+
__m512i t0, t1, t2, t3, t4, t5, t6, t7;
|
437
|
+
__m512i r0, r1, r2, r3, r4, r5, r6, r7;
|
438
|
+
|
439
|
+
initializeState(_);
|
440
|
+
|
441
|
+
index = LOAD8_32(7*(chunkSize / 8), 6*(chunkSize / 8), 5*(chunkSize / 8), 4*(chunkSize / 8), 3*(chunkSize / 8), 2*(chunkSize / 8), 1*(chunkSize / 8), 0*(chunkSize / 8));
|
442
|
+
for(j = 0; j < (chunkSize - rateInBytes); j += rateInBytes) {
|
443
|
+
XORdata21(_, index, (const uint64_t *)input);
|
444
|
+
rounds12
|
445
|
+
input += rateInBytes;
|
446
|
+
}
|
447
|
+
|
448
|
+
XORdata16(_, index, (const uint64_t *)input);
|
449
|
+
XOReq(_me, CONST_64(0x0BULL));
|
450
|
+
XOReq(_sa, CONST_64(0x8000000000000000ULL));
|
451
|
+
rounds12
|
452
|
+
|
453
|
+
index = LOAD8_32(7*4, 6*4, 5*4, 4*4, 3*4, 2*4, 1*4, 0*4);
|
454
|
+
STORE_SCATTER8_64(outputAsLanes+0, index, _ba);
|
455
|
+
STORE_SCATTER8_64(outputAsLanes+1, index, _be);
|
456
|
+
STORE_SCATTER8_64(outputAsLanes+2, index, _bi);
|
457
|
+
STORE_SCATTER8_64(outputAsLanes+3, index, _bo);
|
458
|
+
}
|